
Loguru is a library which aims to bring enjoyable logging in Python.
Did you ever feel lazy about configuring a logger and used print() instead?... I did, yet logging is fundamental to every application and eases the process of debugging. Using Loguru you have no excuse not to use logging from the start, this is as simple as from loguru import logger.
Have you ever struggled with large amounts of geospatial data, huge volumes of files and many custom formats? Have you spent hours — even days — converting and wrangling disparate data formats and wondering how to combine data from different sources? Then this talk is for you!
The solution? Forget about files or force-fitting geospatial data into tabular databases. Imagine a solution that naturally shape-shifts to the underlying data structure. TileDB is this solution: a multimodal database based on multi-dimensional arrays with which you can model any data type. TileDB is architected around a storage engine that uses arrays to store any data type, morphs into specialized analysis applications, supports a range of indexing options, and features an analysis-ready format designed for cloud object storage.
TileDB supports all geospatial data in a unified way with numerous APIs and integrates well with compute and visualization tools. TileDB has integrations with many tools that already exist within Python, such as Dask, Xarray, pandas, PDAL and GDAL. We also build interactive visualization tools with the BabylonJS gaming engine that streams geospatial data directly from TileDB arrays. This makes TileDB a natural fit for geospatial datasets!
When all your data fits naturally into your database's underlying data structures, it becomes much easier to work with. In this talk I will show examples of how to efficiently work with very large geospatial datasets. I will cover how to ingest, load, analyze and visualize all types of geospatial data and show how to combine and use them together.
Gamma-rays are high energy electromagnetic radiation, mainly produced in space, in objects like supernova explosions or black holes. Astrophysicists observe these celestial bodies, to try and understand how gamma-rays are produced in or around them.
Gamma-rays penetrate most materials, they can’t be observed with traditional telescope lenses, that focus light just like a regular camera would. In order to observe gamma rays, astronomers need to use large radiation detectors with thousands (or tens of thousands!) of pixels. The sources of gamma-rays can then be located by tracing back detections in each single pixel to the regions of the sky that could have generated its emission as the telescope is scanning the universe.
This is an intricate problem which requires considerable processing power. In this talk I will discuss how I approached this problem using the NVIDIA CUDA framework to access the computing power of graphical processing units (GPUs) and handle multiple detections at the same time.